Physicians consider benefits, pitfalls of at-home monitoring for wet AMD

Key takeaways:

• One presenter said at-home OCT has the potential to improve outcomes in neovascular AMD through personalized care.
• Another said there are still barriers to overcome with monitoring programs.

SEVILLE, Spain — Home monitoring of neovascular age-related macular degeneration, an interesting concept with many potential benefits, has entered the retina space in the U.S. with the Scanly Home OCT monitoring program from Notal Vision.

But are patients, physicians and health care systems ready for this paradigm shift in the management of neovascular AMD? Two specialists offered their perspectives at the Congress on Controversies in Ophthalmology.

For

Home OCT has the potential to improve outcomes in neovascular AMD, providing personalized patient care, according to Nancy M. Holekamp, MD. Her personal experience in the clinic and authorship in a prospective trial recently published in Retina convinced her that this is the way patients will be managed in the future.

She explained that upon request of the treating physician, the patient receives the OCT monitoring device with instructions for operation. The patient self-images at home on a daily basis, and the data are sent to the cloud, where an AI algorithm reads the images for hyporeflective spaces that correspond to either intraretinal or subretinal fluid. If abnormalities are detected, an alert is triggered, and both the physician and the patient are informed by a message that a visit to the clinic is required.

Nancy M. Holekamp

The retina specialist, however, remains “the captain of the ship,” Holekamp said. It is up to the physician to set the threshold values that trigger the alert. (Holekamp said her threshold is 10 nL of intraretinal or subretinal fluid.) At any time, the health care provider can enter the cloud to view the scans and see how the patient is doing. Once the patient presents for the visit, the specialist reviews the scans and makes the decision on whether to treat.

Holekamp’s prospective study included 17 eyes of 15 patients diagnosed with wet AMD, “but the beauty of this technology is that you can monitor both eyes,” she said.

The patients scanned themselves daily for 6 months and were seen by their physician only when they received the alert. Before home monitoring, the average interval between injections was 8 weeks. During the study time, it was extended to 15.3 weeks.

“The average treatment interval was almost doubled … and we did it without losing any vision. The change in vision was zero letters,” Holekamp said.

Adherence was good: On average, patients performed the scan six times per week, and false alarms were not found to be a concern.

“Out of 40 triggered visits, 39 of them actually had fluid on the OCT, and 37 of them actually received an injection, and the others didn’t based on investigator discretion and tolerance of subretinal fluid,” Holekamp said.

Holekamp presented the case of a 76-year-old patient who was treated with ranibizumab injections followed by the port delivery system (PDS, Roche) in the left eye; he also had a hint of intraretinal cystic change in the right eye. He began home monitoring in December 2022, and in February 2023, the AI algorithm detected intraretinal fluid exceeding 10 nL of volume. The patient was seen, and the PDS was refilled.

“The fluid goes away by the next day, by the next time the patient tests. So, we just keep monitoring this patient, and because he has continuous delivery, he’s doing very well until we see fluctuations in intraretinal fluid. And look at what we never saw before: You can get an accumulation of intraretinal fluid, and then it goes away and then it returns and then it goes away again. We didn’t know that was happening until we had daily monitoring. The amount of data is so instructive here,” she said. “We did a refill exchange again, and the fluid was promptly resolved.”

Home monitoring in this case was also an invaluable means for following up the intraretinal cystic change in the fellow eye.

“We see little tiny changes that are below 10 nL … it ebbs and flows, and we keep following the patient until we see an increase in the fluid that exceeds 10 nL. Then we call the patient, give an injection of faricimab and see the rapid resolution of the intraretinal fluid with a single injection,” Holekamp said. “We just keep following the patient, and the patient does not need another injection until 7.5 months later.”

Home OCT offers “truly personalized patient care” that can decrease treatment burden, Holekamp said.

“What I’ve come to appreciate is that we’ve really been managing our wet AMD patients in the dark with a paucity of data points,” she said. “This is our limited experience in 15 patients and 17 eyes. But right now, in the United States, there’s a much larger clinical trial going on, the DRCR Network Protocol AO. Keep your antenna out for that.”

Against

Home monitoring is a promising concept, but the real-world outcomes are not significantly better than the standard of care to justify the increased cost and physician burden, the organizational problems, the stress of false alarms, and the limitations of the technology in terms of image quality, according to Rodrigo Abreu González, MD, PhD.

In the workflow described by Holekamp, there can be failures and delays at almost every point, he said.

He juxtaposed myths and realities surrounding home monitoring.

Rodrigo Abreu González

“The myth is that with this type of home monitoring, we can detect early and prevent vision loss, but the reality is that visual acuity still declines despite the monitoring,” Abreu González said. “Another myth is the high patient engagement, but in the study, there were only 15 patients, and real-life evidence is inconsistent. The third myth is that home monitoring reduces the burden of clinics. What really happens is that it increases the alerts and workflow stress and not only for the doctors. Finally, the myth is that AI ensures safety and precision, but it still needs manual review and has false alerts.”

The ALOFT study (Mathai et al.) analyzed long-term outcomes of the AI-enabled ForeseeHome AMD monitoring program (Notal Vision). The percentage of conversions to wet AMD detected by the system was 52%, only slightly above the percentage detected by routine visits.

When comparing the outcomes of home monitoring with the standard of care, Abreu González said that visual outcomes were better with home monitoring, with a mean loss of four letters on average compared with 11 and 82% of patients maintaining 20/40 or better vision as compared with 62%.

“However, there was still a loss of vision, and 20% of the patients lost vision below 20/40,” he said.

One important drawback of at-home OCT technology is the inferior image quality compared with the high standards of OCT in the clinic. Another pitfall is the need for technical support, training and follow-up.

“We have patients who are unable to use a smartphone, laptop or iPad, so just imagine if they are using the OCT at home,” he said.

Stress and burden are not reduced but multiplied, according to Abreu González. Retinal fluid has fluctuations that do not necessarily indicate an activity that requires treatment. But these fluctuations might trigger an alert that generates anxiety in the patient and extra work for the physician.

In addition to the patients the physician is taking care of in the clinic, there are alerts and messages from the patients at home. This leads to work overload and increased infrastructure cost. Depending on the country and the system, there may also be reimbursement challenges, Abreu González said.

He said that home monitoring systems are, at present, not mature enough to be implemented as there are still too many patient-, technology- and provider-related barriers that need to be overcome.

Recent data showed the weaknesses of all currently available home monitoring technologies.

“Current home vision tests lack sufficient diagnostic accuracy (Hogg et al.), almost 30% of the patients cannot complete the tests without assistance (O’Connor et al.), and AI-generated alerts still require manual review (Willis et al.),” Abreu González said.

Better real-world evidence from independent long-term studies as well as simplified, patient-friendly devices are needed, he said.

References:

• Hogg RE, et al. JAMA Ophthalmol. 2024;doi:10.1001/jamaophthalmol.2024.0918.
• Holekamp NM, et al. Retina. 2024;doi:10.1097/IAE.0000000000004167.
• Mathai M, et al. Ophthalmol Retina. 2022;doi:10.1016/j.oret.2022.04.016.
• O’Connor SR, et al. Int J Environ Res Public Health. 2022;doi:10.3390/ijerph192013714.
• Willis ET, et al. Ophthalmol Ther. 2024;doi:10.1007/s40123-024-00953-8.

Published by:

Sources/Disclosures

Collapse

Source:

Holekamp N, et al. Debate 2: Home monitoring has the potential to improve patient outcomes in AMD. Presented at: Congress on Controversies in Ophthalmology; April 4-5, 2025; Seville, Spain.

Disclosures:
Abreu González reports no relevant financial disclosures. Holekamp reports being an employee of Roche and holding stocks or stock options in Nacuity, Notal Vision and Roche.

Read more about